**** BEGIN LOGGING AT Tue Jun 24 02:59:59 2014 Jun 24 03:00:28 nothing bad about brief, as long as it's clear that Pseudo-Dual and Dual-SIM-Dual-Standby are two completely different links Jun 24 03:00:34 yup Jun 24 03:00:52 things** Jun 24 03:01:05 s/links/things/ Jun 24 03:01:40 How's this one? Jun 24 03:01:42 "Dual standby phones allows both SIMs to be accessed through [[time multiplexing]], but only one of the SIMs can be used at a time when making calls." Jun 24 03:01:55 http://talk.maemo.org/showthread.php?p=1430792#post1430792 might explain a tad better Jun 24 03:02:20 yes, that's perfect Jun 24 03:03:13 Basically, DSDS works with the modem listening into both networks on just one receiver Jun 24 03:03:25 right? Jun 24 03:03:28 right Jun 24 03:04:37 modern modems always sleep for a few seconds, then listen for a short while if an inbound call gets signalled, then go asleep again Jun 24 03:05:06 obviously during sleep period the modem can listen to another band (or even several others) Jun 24 03:05:19 sorta like multitasking Jun 24 03:05:25 yes, exactly same Jun 24 03:05:54 with same requirements to save context of one task and restore context of the other one Jun 24 03:06:15 thus hte modem needs sufficient RAM to do DSDS Jun 24 03:06:28 figures Jun 24 03:06:45 but it neeeds only one RX/TX Jun 24 03:07:18 what about the MTK-based phones, i.e. the MT6572? Jun 24 03:07:31 SIM is inactive and even powered down for hours, before modem needs it again Jun 24 03:07:45 the MTK does DSDS Jun 24 03:07:45 The Lenovo I'm using right now is Dual SIM, but there's no mention of what tech it uses Jun 24 03:07:54 Ah noted Jun 24 03:08:27 the modems we initially planned to use for Neo900 were MTK based and offered dual SIM Jun 24 03:08:49 GM601/801, iirc Jun 24 03:09:25 only prob is with Mediatek being an ass when it comes to open-source Jun 24 03:09:32 from what I heard Jun 24 03:09:50 well, for the modem there's no chance for ever seing sth like open source Jun 24 03:09:58 seeing* Jun 24 03:10:07 So projects like Replicant are basically a pipe dream Jun 24 03:10:16 the radio stack is closed by legal requirements even Jun 24 03:10:30 no, since replicant doesn't run on the modem Jun 24 03:10:46 They had more than enough trouble with dealing with GPU drivers Jun 24 03:11:04 we have a Linux CPU that's completely detached (except for a data link, e.g. USB) from modem Jun 24 03:11:16 DocScrutinizer05: I thought it was just legal paranoia? No court rulings on that, are there? Jun 24 03:11:43 well, the modem needs cert. The cert includes the firmware/software Jun 24 03:11:54 changing software voids certification Jun 24 03:11:56 DocScrutinizer05: sure, but no reason AFAIK the source for that can't be released Jun 24 03:12:03 you're operating an illegal device Jun 24 03:12:09 if you make use of it, yes Jun 24 03:12:11 Does anyone actually care, once you've got it in the user's hands? Jun 24 03:12:28 I mean... unless I turn it into a jammer... Jun 24 03:12:37 (although I imagine the laws have some exceptions for testing/development…?) Jun 24 03:12:38 yes, when a phone is known to usually get abused, you *will* be in trouble Jun 24 03:13:17 Luke-Jr: actually not even then. Even in lab you *should* ensure that your DUT doesn't emit RF Jun 24 03:13:34 Reminds me of the issues with AMPS phones Jun 24 03:13:36 DocScrutinizer05: sure, I mean exceptions to the cert Jun 24 03:13:45 in the lab, you won't have a cert yet Jun 24 03:13:47 when you go testing in the field, you coordinate with carier about it Jun 24 03:14:24 My old man was quite pissed off when he found out that his cellphone line was being tapped for fraudulent phone calls Jun 24 03:14:36 hehe Jun 24 03:14:40 too bad government is so .. slow. seems like they should just do digital signatures on firmware images themselves ;) Jun 24 03:15:00 "to flash this within US territory, wait for a GPS lock, then upload firmware signed by " Jun 24 03:15:25 Some asshole tapped into his phone, placed calls on different locations, and left him with a hefty postpaid bill Jun 24 03:15:27 the firmwares are signed and thus no way to tamper with them. But that's part of modem manuf's responsibility Jun 24 03:15:54 Luke-Jr: see Jun 24 03:16:04 ~wiki tivoization Jun 24 03:16:07 At http://en.wikipedia.org/wiki/Tivoization (URL), Wikipedia explains: "'Tivoization' {{IPAc-en|ˈ|t|iː|v|oʊ|ɨ|ˌ|z|eɪ|ʃ|ən}} is the creation of a system that incorporates software under the terms of a copyleft software license (like the GPL), but uses hardware restrictions to prevent users from running modified versions of the software on that hardware. Richard Stallman coined the term in reference to TiVo's use of GNU GPL licensed softwa Jun 24 03:16:37 DocScrutinizer05: right, my point is such things, while perhaps GPL-incompatible, don't prevent open sourcing under some less protective license Jun 24 03:16:46 eg, perhaps BSD or something Jun 24 03:17:04 another aspect is that there are literally 1000 man years in such a firmware, and that's an asset the company won't disclose as FOSS Jun 24 03:17:40 meh, that's the same old excuse used for any proprietary sw Jun 24 03:17:46 sure Jun 24 03:18:06 how about this? Jun 24 03:18:17 Dual standby phones allows both SIMs to be accessed through time multiplexing. When making or receiving calls, the modem locks to the active channel... Jun 24 03:18:30 but you won't find a way to convince Ericsson or Siemens/Gemalto or whomever to go FOSS with their radio stack Jun 24 03:18:44 DocScrutinizer05: isn't there an open GSM stack? Jun 24 03:18:50 Guest93593: perfect Jun 24 03:19:05 Luke-Jr: OsmocomBB Jun 24 03:19:31 GSM only, and there not even EDGE Jun 24 03:19:39 Mind if you guys provide some further explanation on what I wrote? Jun 24 03:19:48 and only available for TI Calypso Jun 24 03:19:50 does it run on the same modems that could run a non-free 3/4G stack? Jun 24 03:20:08 ^^^ Jun 24 03:20:14 * Luke-Jr looks it up Jun 24 03:20:35 Guest93593: what do you want to know? Jun 24 03:21:01 Modem locks to the active channel when a call is made, but what's next? Jun 24 03:21:17 maybe missing: "...while the other cannel is "not available" during that time" Jun 24 03:21:53 s/channel is/channel gets ignored and thus is/ Jun 24 03:22:02 s/cannel is/channel gets ignored and thus is/ Jun 24 03:22:02 DocScrutinizer05 meant: maybe missing: "...while the other channel gets ignored and thus is "not available" during that time" Jun 24 03:22:50 next: the call proceeds. After call terminates, the modem resumes scanning both channels Jun 24 03:23:18 "Dual standby phones allows both SIMs to be accessed through time multiplexing. When making or receiving calls, the modem locks to the active channel; the other channel would be ignored and thus unavailable during the duration of the call." Jun 24 03:23:19 pretty simple logic Jun 24 03:23:42 yes. exactly Jun 24 03:24:09 That's 'swell Jun 24 03:24:49 maybe a sidenote: hefty data traffic and a call are basically same for a modem Jun 24 03:25:11 but the modem can interrupt data, it cannot interrupt a call Jun 24 03:25:16 yup Jun 24 03:25:20 "Dual standby phones, such as those running on Mediatek chipsets..." Jun 24 03:25:23 depends on implementation Jun 24 03:26:57 * DocScrutinizer05 has to get some food, and a nice coffee. alas this means leaving the flat and facing big bluebox. bbl Jun 24 03:27:11 cya dude Jun 24 03:27:14 o/ Jun 24 03:27:18 well I have to eat sume lunch anyway Jun 24 03:34:36 ~flasher Jun 24 03:34:36 extra, extra, read all about it, flasher is at http://www.jedge.com/n810/flasher/maemo_flasher-3.5_2.5.2.2.tar.gz (also .exe!), or http://www.chakra-project.org/ccr/packages.php?ID=5027 or generally http://www.google.com/search?q=maemo_flasher-3.5_2.5.2.2.tar.gz. HARMattan(N9): https://aur.archlinux.org/packages/fl/flasher-harmattan/flasher-harmattan.tar.gz, or -- list of filenames/md5sums: http://pastebin.com/sYKdNJSH, or http://galif.eu/nokia/ Jun 24 03:38:08 ~combined Jun 24 03:38:08 rumour has it, combined is the rootfs fiasco image of maemo. For N900 latest (PR1.3.1) see http://nds2.fds-fire.nokia.com/fdp/interface/FiRe/2011/9/--FID--A0A22MVWFVFAM/--LID--FiRe1317015685654/RX-51_2009SE_21.2011.38-1_PR_COMBINED_MR0_ARM.bin, or http://galif.eu/nokia/ Jun 24 03:38:23 ~emmc Jun 24 03:38:23 [emmc] magnet:?xt=urn:btih:402fb5cc8a48ecbc18a77c9cf70d869a775bcf53&dn=RX-51%5F2009SE%5F10.2010.13-2.VANILLA%5FPR%5FEMMC%5FMR0%5FARM.bin or https://dl.dropboxusercontent.com/u/4105928/RX-51_2009SE_10.2010.13-2.VANILLA_PR_EMMC_MR0_ARM.torrent, or http://nds2.fds-fire.nokia.com/fdp/interface/FiRe/2010/5/--FID--A0A22YHFSICNA/--LID--FiRe1275051276916/AE98ED9D_RX-51_2009SE_10.2010.13-2.CENTRAL-EUROPE_PR_EMMC_MR0_ARM.bin Jun 24 03:39:16 thanks. ^^ Jun 24 03:39:21 ~sb Jun 24 03:39:21 methinks scratchbox is a cross-compiling system that uses binfmt_misc, rpc calls, and an nfs mount to make a cross-build appear to be 100% native, and is found at http://www.scratchbox.org/, hosted by maemo now. Also at http://maemo.merlin1991.at/files/SB Jun 24 03:39:22 yw :-9 Jun 24 03:39:27 :-) Jun 24 13:26:14 DocScrutinizer05: Thanks, so do they have a set length or order number I need to look for. Jun 24 13:26:36 Maxdamantus: Which version of CSSU do you have the issue with rotation. Jun 24 14:04:37 sixwheeledbeast: hm. I'm not sure how to determine what version it is. Jun 24 14:05:50 The problem's fairly simple though: I disabled the rotation (iirc, using gconf) because I often have a bunch of browser windows left open, and it takes a while to rotate them (probably moving them out of and back into swap), and I always just used it in landscape anyway. Jun 24 14:06:39 But the phone application rotated still, as it does without CSSU, but it also rotates the browser windows in the background. Jun 24 14:07:39 Dunno if the normal Nokia distribution rotates windows in the background when using the phone application. Jun 24 14:08:17 Maxdamantus: you should then lock the phone application Jun 24 14:08:31 Maxdamantus: that's how it was designed to be Jun 24 14:08:50 call-ui has it's own rotation settings Jun 24 14:08:55 I just set the phone application to landscape mode (it's a setting within the application—dunno if it's CSSU) Jun 24 14:09:00 Yeah. Jun 24 14:09:26 arcean: how what was designed to be? Jun 24 14:09:32 Has that "fixed your issue Jun 24 14:09:38 i was talking about "Dunno if the normal Nokia distribution rotates windows in the background when using the phone application." Jun 24 14:10:11 arcean: seems like a weird thing to design it to do, seeing as they'll always rotate back to landscape before you manage to focus them. Jun 24 14:10:20 (in the Nokia distribution) Jun 24 14:10:59 #maemo getversion Jun 24 14:11:04 ~getversion Jun 24 14:11:04 getversion is probably "input the following in a terminal 'osso-product-info | grep OSSO_VERSION | cut -d "'" -f 2; dpkg-query -W -f='${Version} ' mp-fremantle-community-pr; uname -r'" Jun 24 14:11:27 Maxdamantus: ^^ can you get the output of above please. Jun 24 14:11:44 Jun 24 14:12:05 The "Device info" or whatever it is in Settings says something like "(null") Jun 24 14:12:31 mmm Jun 24 14:13:33 * Maxdamantus wonders where it gets the information. Jun 24 14:13:48 `strace -e open -f osso-product-info` only shows it opening .sos. Jun 24 14:14:02 (and ld.so.cache) Jun 24 14:14:03 so the output of the above command is with nothing after it? Jun 24 14:14:41 I'd at least expect the dkpg-query to work Jun 24 14:14:48 Oh, I didn't run the dpkg-query bit. Jun 24 14:15:00 the whole thing please Jun 24 14:15:44 don't include the 'quotes around the command' Jun 24 14:17:05 It looks like there's an extra single quote in that command btw Jun 24 14:17:39 21.2011.38-1Tmaemo9.1 2.6.28.10-cssu1 Jun 24 14:29:46 Maxdamantus: so that's your version, thumb and no kp Jun 24 14:30:56 It looks like there is but there isn't an extra quote, it's the cut command Jun 24 14:31:20 Actually, I was thinking about the: .. uname -r'" Jun 24 14:31:34 but now I see that there's both a double quote and a single quote at th ebeginning. Jun 24 14:31:51 "input the following in a terminal '.. Jun 24 14:32:22 yep, I think it maybe a good idea to remove the quotes around things. Jun 24 14:32:40 it could be confusing Jun 24 14:34:04 no, infobot #maemo getversion is input the following in a terminal: osso-product-info | grep OSSO_VERSION | cut -d "'" -f 2; dpkg-query -W -f='${Version} ' mp-fremantle-community-pr; uname -r Jun 24 14:34:46 ~getversion Jun 24 14:34:47 somebody said getversion was "input the following in a terminal 'osso-product-info | grep OSSO_VERSION | cut -d "'" -f 2; dpkg-query -W -f='${Version} ' mp-fremantle-community-pr; uname -r'" Jun 24 14:36:01 ~getversion =~ s/"input/input/ Jun 24 14:36:01 Maxdamantus: i didn't have anything called 'getversion' to modify Jun 24 14:36:14 ~#maemo getversion =~ s/"input/input/ Jun 24 14:36:14 Maxdamantus: OK Jun 24 14:36:26 ~#maemo getversion =~ s/r'"/r'/ Jun 24 14:36:26 Maxdamantus: OK Jun 24 14:36:43 ~#maemo getversion =~ s/ 'osso/: osso/ Jun 24 14:36:44 Maxdamantus: OK Jun 24 14:36:48 ~getversion Jun 24 14:36:48 i guess getversion is input the following in a terminal: osso-product-info | grep OSSO_VERSION | cut -d "'" -f 2; dpkg-query -W -f='${Version} ' mp-fremantle-community-pr; uname -r' Jun 24 14:36:58 er, one more. Jun 24 14:37:06 ~#maemo getversion =~ s/r'/r/ Jun 24 14:37:07 that doesn't contain 'r'', Maxdamantus Jun 24 14:37:14 ~getversion Jun 24 14:37:14 rumour has it, getversion is input the following in a terminal: osso-product-info | grep OSSO_VERSION | cut -d "'" -f 2; dpkg-query -W -f='${Version} ' mp-fremantle-community-pr; uname -r Jun 24 14:37:19 done Jun 24 14:37:30 Ah. Jun 24 14:37:57 missed the ~ the first time Jun 24 14:38:47 Anyway Maxdamantus is your rotation issue fixed? Jun 24 14:39:30 I wanted to see if you where on CSSU-S which still has a rotation bug in hildon-desktop. Jun 24 14:39:47 However you are not. Jun 24 14:40:08 Depends what you mean by fixed. Jun 24 14:40:29 It shouldn't arise for me anymore, because I set the phone application to stay in landscape mode. Jun 24 14:41:03 but the software won't've changed. If I set it back to portrait, I suspect it'll still rotate the windows in the background. Jun 24 14:42:21 I see, will I think what you describe is correct if I have understood correctly. Jun 24 14:42:24 (note: I also configured the desktop not to switch between portrait/landscope generally) Jun 24 14:42:28 s/will/well/ Jun 24 14:42:28 sixwheeledbeast meant: I see, well I think what you describe is correct if I have understood correctly. Jun 24 14:43:42 So with the phone application in portrait mode, if I exit it or go to the window switcher thing, it'll rotate back to landscape. Jun 24 14:45:10 the phone application and browser has there own settings everything else will follow hildon-desktop if there rotation compatible. Jun 24 14:46:16 So yes to your question if you have desktop set to landscape only. Jun 24 14:46:30 Except the other applications follow the phone application, but unless the system's lagging, you don't notice because it switches back to landscape as you switch to the other applications. Jun 24 14:53:36 sixwheeledbeast: http://wiki.maemo.org/User:Joerg_rw/tools#no_autorotating Jun 24 15:06:09 Yes locking h-d from rotating browser and phone have separate settings. Jun 24 15:07:37 * sixwheeledbeast boots his testing N900 Jun 24 15:15:34 Right I understand. But this is a none issue really. You shouldn't see the other applications as pressing task switcher, switches everything to landscape again before there shown. You would have to have weird transitions settings to even notice. Jun 24 15:17:00 or lots of browser windows open. Jun 24 15:17:37 The problem was that the browser windows caused the system to lag when someone called, so I couldn't answer it. Jun 24 15:17:54 someone calls, phone application opens, twenty web pages begin to rotate. Jun 24 15:29:04 Ah, I see. Well stock Maemo wouldn't have done this as desktop ui never rotated before. Jun 24 15:32:40 Not even the phone application? Jun 24 15:33:04 or are you saying that applications don't rotate in the background there? Jun 24 15:41:46 in the background, the phone application rotated Jun 24 15:42:54 general advice: never keep many microB windows open, always close a window after you finished with what you opened it for Jun 24 15:43:33 * Maxdamantus doesn't often finish what he opens web pages for. Jun 24 15:43:58 * Maxdamantus has 361 tabs open on his computer. Jun 24 15:44:19 and 13 PDFs open on one desktop. Jun 24 15:45:45 sure, me too. But I hardly ever open a browser window on N900 Jun 24 15:46:41 and when I do, I don't switch to other tasks, rather I close the browser window to go to other task Jun 24 15:47:11 browser windows are memory (and cpu) hogs by design Jun 24 15:55:31 I think by design, they're not in browserd. Jun 24 15:55:56 It looks like they forget everything after a while and just keep a picture of the page. Jun 24 15:56:17 So when you focus the window again it reloads the page and jumps to where you were. Jun 24 15:57:06 or when the window gets rotated. Jun 24 16:04:31 Difference here is tabs are minimized on your desktop however in microVB there is a window for every one. Not good for resources even on thumb Jun 24 16:06:24 I would seriously hope that browser pages are not continuously redrawn while you are doing something else Jun 24 16:11:38 hmm Jun 24 16:12:14 you for sure have a canvas for each browser window/tab Jun 24 16:12:56 I don't think windows vs. tabs should make much of a difference? Jun 24 16:12:58 that canvas is even larger on desktop than in microB Jun 24 16:13:15 I suspect most people have at least 30 X windows open on their N900 normally anyway. Jun 24 16:13:33 hmm ... I dont Jun 24 16:13:34 Every widget and icon on the desktop is a window. Jun 24 16:13:37 thus no, tabs vs windows makes no diff, unless you're using FF which rneders tabs only when you activate them Jun 24 16:13:38 (same as Doc) Jun 24 16:14:11 * Maxdamantus doesn't, because he doesn't have much on his desktop. Jun 24 16:14:44 * Maxdamantus has four icons. Jun 24 16:14:56 a desktop icon is maybe 50*50, a microB canvas prolly 1000*1000 Jun 24 16:16:14 anyway my swap usage is usually pretty moderate or even humble Jun 24 16:16:22 Are you thinking about resources the compositor has to deal with? Jun 24 16:16:39 and I never suffer the "slows down after 3 days" swap fragmentation issue Jun 24 16:16:48 since nothing swaps, usually Jun 24 16:17:15 who cares abozut compositor? Jun 24 16:17:25 ? Jun 24 16:17:52 go fullscreen and no compositor involved at all Jun 24 16:18:00 :) Jun 24 16:18:06 is what they told me Jun 24 16:18:38 sill a 20 microB windows will kill your swap Jun 24 16:18:41 still* Jun 24 16:19:12 since every single one has a canvas (prolly from X) that's pretty huge Jun 24 16:19:36 several MB Jun 24 16:20:12 I suspect most people have at least 30 X windows open on their N900 normally anyway. Jun 24 16:20:14 this is me Jun 24 16:20:25 and I hate microB's lack of tabs Jun 24 16:20:33 *was Jun 24 16:20:40 I don't think X maintains a buffer for stuff that's not visible. Jun 24 16:20:41 *hated Jun 24 16:20:46 but compositors normally do. Jun 24 16:20:49 meh, I can open 100 windows and device still snappy. When those windows are not greedy Jun 24 16:20:55 well for tabbed browsing try Opera Jun 24 16:21:22 Maxdamantus: the browser does Jun 24 16:21:34 Yes, the browser seems to. Jun 24 16:21:42 either that or the compositor does. Jun 24 16:22:06 It could just be the compositor .. seems like a sensible optimisation. Jun 24 16:22:49 again: those who should know told us that compositor is only used e.g. for the title bar/systray Jun 24 16:24:25 iirc Jun 24 16:25:44 So how does it make the small pictures of windows? Jun 24 16:25:53 in the task switcher Jun 24 16:25:55 DocScrutinizer05: you mean you tested with 100 xterm ? ;) Jun 24 16:26:53 * Maxdamantus wrote a command here to launch lots of xterms with random colours once. Jun 24 16:37:12 bencoh: exactly. well, almost Jun 24 16:38:24 there been a news on the nets a 4 or 5 years ago: "XY(some android crap?) runs 12 tasks concurrently!" We went MUHAHAHABWAHA here and somebody opened 117(iirc) windows on N900 Jun 24 16:39:01 :) Jun 24 16:50:51 haha Jun 24 16:51:40 the fact that N900 had swap in additional to its' 256MB RAM really helped back when high-end *was* 256MB RAM Jun 24 16:54:12 * Maxdamantus is running his IRC client on a machine with the same amount. Jun 24 17:11:02 hah Jun 24 17:11:20 the client runs on my laptop (16gb) and the bouncer runs on a server with 32gb Jun 24 17:11:28 * kerio never swaps Jun 24 17:13:54 my server only has 512mb and the secondary server has 4gb Jun 24 17:14:24 512 millibits isn't much. Jun 24 17:14:50 main disk is 2gb Jun 24 17:15:07 and there is a 32gb stick too Jun 24 17:15:08 Not sure what that is. Two gram bits? Jun 24 17:15:21 Though that should be written g b, not gb Jun 24 17:15:28 gigabytes and megabytes Jun 24 17:15:35 woosh Jun 24 17:15:37 Oh, GB, MB Jun 24 17:15:59 Maxdamantus, the SI fetishist Jun 24 17:16:02 yes, I know.. shift is so far away on this keyboard Jun 24 17:16:08 it should be GiB anyway Jun 24 17:16:11 Though I suspect you're actually talking about gibibytes and mebibytes, since they're RAM sizes. Jun 24 17:16:31 i wonder why ram producers didn't go the hard disk producers' route Jun 24 17:16:52 Heh. Jun 24 17:17:25 * Maxdamantus has 34.359738368 GB of RAM in his current machine. Jun 24 17:17:55 i have 17.179869184 GB Jun 24 17:18:42 and my stupid windows 7 says 21GB Jun 24 17:18:52 u wot m8 Jun 24 17:19:17 He Windows 7 mate. Jun 24 17:19:42 (actually 8GB but this has too many hard drives with paging/swap) Jun 24 17:21:37 Dammit need more ram on my VM server. It only has 6GB RAM... Jun 24 17:21:54 mine is a dedicated server Jun 24 17:21:57 and i don't even pay for it Jun 24 17:22:05 I think you mean 8.589934592 GB Jun 24 17:22:20 (aka 8 GiB) Jun 24 17:22:29 who are you to decide how much ram he has? Jun 24 17:22:42 I'm just guessing. Jun 24 17:22:51 maybe he has 5.58GiB of ram Jun 24 17:23:07 Oh, he said 6. Jun 24 17:23:19 I think he means 6.442450944 GB then. Jun 24 17:23:30 you've been already proven wrong at least once Jun 24 17:28:23 With a VM, it's not unreasonable to have 6.0GB of RAM... Jun 24 17:29:22 Depends what you consider to be RAM. Jun 24 17:29:30 I wouldn't consider it to be dependent on the software. Jun 24 17:30:00 Some of my RAM is used by the onboard graphics card .. doesn't mean I have any less RAM. Jun 24 17:30:39 but if you can't use more than the amount of ram the hypervisor limited you to... then you have exactly that much... Jun 24 17:31:04 That's a fairly arbitrary delimitation. Jun 24 17:31:18 Why not exclude the memory used by the kernel, since you can't use that either? Jun 24 17:31:29 you are using the kernel memory Jun 24 17:31:40 No. The kernel is using the kernel memory. Jun 24 17:31:59 but it's your kernel, reduce the size of the kernel if you want more user memory Jun 24 17:31:59 Just as the hypervisor is using the hypervisor memory that you can't see if you're in a VM. Jun 24 17:32:06 and the memory used by other VM. Jun 24 17:32:09 VMs* Jun 24 17:32:27 So if I run a VM, it's not my VM? Jun 24 17:32:34 but if I run a kernel, that's my kernel? Jun 24 17:32:34 hypervisor memory isn't your memory. Jun 24 17:32:46 Sure it is, if I own the machine. Jun 24 17:32:51 it's not if you don't. Jun 24 17:32:59 If you don't own the machine, you don't have any of its RAM. Jun 24 17:33:11 then kerio doesn't have any memory. Jun 24 17:33:38 your mom doesn't have any memory Jun 24 17:38:23 i think his point is that ram has to be a physical chipset Jun 24 17:41:56 and that delimiting it to a "VM" is just as arbitrary as delimiting it to anything else you happen to have access to. Jun 24 17:43:33 * Maxdamantus sleeps. Jun 24 17:45:43 You have 16GiB RAM when you rented your VM for 4GB, because the host has 16GiB. Jun 24 20:00:07 ~ping Jun 24 20:00:08 ~pong **** ENDING LOGGING AT Wed Jun 25 02:59:58 2014